Journal of Investment Strategies

Risk.net

Real-time prediction and post-mortem analysis of the Shanghai 2015 stock market bubble and crash

Didier Sornette, Guilherme Demos, Qun Zhang, Peter Cauwels and Qunzhi Zhang

  • Real-time predictions of the automated analysis of the Financial Crisis Observatory at ETH Zurich
  • Analysis of the bubble regime in Chinese stock markets since mid-2014 using the LPPLS model that focuses on non-sustainable super-exponential log-price behaviour
  • Introduction and presentation of the use of the DS LPPLS Confidence and DS LPPLS Trust index as advanced diagnostics of bubbles
  • Interpretation of the Chinese bubble in analogy with previous Western market bubbles as catalyzed by authorities, suggesting a general trend globally to use the stock markets as the new engines of growth.

ABSTRACT

The authors assess the performance of the real-time diagnostic, available to the public on the website of the Financial Crisis Observatory (FCO) at ETH Zürich, of the bubble regime that began developing in Chinese stock markets in mid-2014 and started to burst in June 2015. The analysis is based on (i) the economic theory of rational expectation bubbles; (ii) the behavioral mechanisms of imitation and the herding of investors and traders; and (iii) the mathematical formulation of the log-periodic power lawsingularity (LPPLS), which describes the critical approach toward a tipping point in complex systems. The authors document how the real-time predictions were presented in the automated analysis of the FCO, as well as in their FCO Cockpit report of June 2015. A complementary post-mortem analysis of the nature and value of the LPPLS methodology in diagnosing the Shanghai Composite Index bubble and its termination is also given.

To continue reading...

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an indvidual account here: